Reference library
Multilayer feedforward networks are universal approximators
Paper summary
Establishes broad approximation results for feedforward networks with suitable nonlinear hidden units. The analysis treats network expressiveness as a property of function classes rather than of a specific training algorithm.
Relevance to the project
A foundational justification for flexible neural mechanism functions. It leaves open how evolution or learning finds a useful, efficient F and updater.
Approximation theorems do not guarantee generalization, stability, or small networks.
Source record
- Reference ID
- article-https-www-sciencedirect-com-science-article-pii-0893608089900208
- Archived version
- Published Neural Networks PDF hosted by CMU course archive
- Summary provenance
- reference-library/papers/article-https-www-sciencedirect-com-science-article-pii-0893608089900208/metadata.json · summary
- Review status
- initial orientation summary; not a critical full-paper review
- Reading scope
- Selected abstract/opening and available closing discussion/conclusion passages in the preserved text extract; broader project notes provide context. No reproduction of experiments, exhaustive methods review, or equation/figure audit.